Search papers, labs, and topics across Lattice.
Tsinghua University
1
0
3
Current vision-language models falter in streaming interaction understanding, with alarming mis-calibration leading to confidently incorrect predictions.
